Document By

SANTOSH BHARADWAJ REDDY

Email: help@matlabcodes.com
Engineeringpapers.blogspot.com
More Papers and Presentations available on above site

ABSTRACT

The issue of Power Quality is of paramount importance in the

distribution and consumption of electric power. There are numerous facts of power
quality disturbances and each has its own source and mitigation techniques. The first step
towards any solution for a disturbance is to recognize the presence of a particular type of
disturbance and locate its source. Conventional methods for recognition of a power
quality disturbance consists of collecting operating data, inspecting the wave forms
visually and then identifying any disturbance that may persist in the data. Although the
available measuring and recording devices offer substantial help, the process is, in the
main, very slow. At the University of Tasmania, a project is under way to automatize
this process. The ultimate goal is to develop an Automatic Disturbance Recognition
System (ADRS). In its current state, tests with simulated as well as real disturbance data
yielded encouraging results, some of which are presented in this paper.

The main purpose of this paper is to review the state of the art techniques
in signal processing for automatic classification of Power quality events and to give a
sign of the next trends. The time when power quality monitoring equipment just takes
pictures of raw waveforms has gone. Nowadays, power quality monitoring systems are
striving to identify and classify events automatically in order to solve problems in
electrical networks. These power quality-monitoring systems are also explained in brief
in this paper.

1. INTRODUCTION:

The term “Power Quality” has been used to describe the extent of
variation of the voltage, current and frequency on the power system. Most power
apparatus made over a decade back could operate normally with relatively wide
variations of these parameters. However, equipment added to power system in recent
years generally is not tolerant to these variations for two main reasons. First one is that
design tolerances have been going down in a competitive market and second one is the
increasing use of sophisticated electronic controls. Hence, system disturbances, which
were tolerated earlier, may now cause interruption to industrial power system with a
resulting loss of production and this could be substantial with greater stress on
productivity & quality now.

Especially for a developing country like India, power quality (PQ) is of

prime importance considering the need for energy conservation. It is a paradox that some
of the energy conserving devices themselves is the reason for some of the power quality
problems. Though the need and importance for improving power quality is well known
and solutions also have been available for some time, the improvements are not taking
place due to certain institutional and technological barriers. This paper analyses these and
proposes an integrated approach to overcome these barriers for effectiveness.

The quality of electric power is becoming a matter of increasing concern to

both power utilities and their customers. There are a number of major reasons for this.
They are:
1.Customer equipment today is more sensitive to power quality variations than equipment
used in the past. Modern microprocessor-based controllers and power electronic devices
are very sensitive to various disturbances in the power supply.
2. Increasing levels of harmonics in power systems give rise to concerns about the future
impact on the system performance.
3.Customers are becoming better informed about power quality problems and are
challenging the utilities to improve power quality.

Power quality may deteriorate due to a variety of causing a particular type of

power quality problem and to locate the sources of that disturbance in the power system
so that corrective action can be taken. Existing methods of analyzing and identifying
power disturbances are laborious since they are based on visual inspection of disturbance
waveforms. Due to the complexity of power quality problems and the lack of reliable
techniques to analyze these problems, power utilities are unable to ensure the required
level of power quality without a considerable increase in cost.

2.WHAT IS POWER QUALITY ?

Power quality or, in fact, lack of quality is a term used to describe the most
important aspect of the electricity supply. Power quality can be defined as any problem
manifested in voltage, current, or frequency deviation that results in failure or mal-
operation of electric equipment. A short description of each individual power quality
problems is given below:
Voltage Sag: A decrease of the RMS voltage to 0.1~0.9pu for a duration of 0.5 cycle to 1
minute
Voltage Swell: An increase of the RMS voltage to 1.1~1.8pu for a duration of 0.5 cycle
to 1 minute.
Momentary Interruption: A decrease of the RMS Voltage to less than 0.1pu, for
duration of 0.5 cycles to 1minute
Voltage Fluctuation: Variation of voltage envelop between 0.9 to 1.1pu.
Capacitor switching transients: A sudden non-power frequency change in the steady
state of Voltage.
Impulse transients: A sudden non-power frequency change in the steady state of
Voltage with duration typically in the order of tens of microsecond.
Harmonics: sinusoidal voltage having frequencies that are integral multiples of power
frequency.
Frequency Deviation: An increase of the power frequency to 104~110% of nominal
frequency or a decrease to 90~96% of nominal frequency.
Notch: A switching disturbance of the normal power voltage waveform, which is of
opposite polarity to the waveform with each notch lasting less than 0.5 cycle.
Noise: Unwanted electrical signals superimposed upon voltage with broadband spectral
content lower than 200kHz.
DC Offset: A dc component of 0~0.1pu superimposed on the RMS voltage.
A normal waveform can, therefore, be visualized as one devoid of any of the above
power quality problems. In essence a normal waveform is a power frequency (50 HZ)
sinusoidal waveform with RMS voltage within 0.95~1.05pu.

3. COSTING OF POOR POWER QUALITY:

A major hurdle in the improvement of power quality is the lack of clear

understanding of the implications of poor power quality and the quantification of effects
of poor power quality. In most cases what is considered as the cost of poor power quality
is just the penalty / tariff to be paid for not meeting the norms for power quality. Though
in principle this is correct, considering the fact that most utility regulations do not
adequately capture the total cost of poor power quality, this leads to a wrong picture.
These include increased productivity, better quality of production,
enhanced reliability and life of equipment, reduced maintenance costs, savings in energy
etc. In many cases the direct benefits may be larger than the savings effected on
electricity.

4. POWER QUALITY MONITORING:

A basic requisite for costing (quantification) of poor power quality and also
for the formulation of proper standards, guidelines & regulations is the measurement of
power quality and the availability of power quality data. PQ variations such as
momentary interruptions, voltage sags, switching transients and harmonic distortion can
impact customer operations, causing equipment damage and significant costs in lost
production and down time.

Need for monitoring & various applications:

PQ monitoring has traditionally been used for problem solving in industrial, commercial
and residential systems. PQ monitoring is becoming an integral part of overall system
performance assessment.
Benefits of PQ monitoring include
Understanding PQ & reliability: The monitoring system permits the utility to
characterize the system performance in terms of its impact on customer equipment.
Customers are increasingly concerned about this as deregulation takes over and there
may be less incentive for the distribution company to provide the levels of PQ that are
required for proper operation of the customer’s facility.
Prioritizing system improvements: Utilities traditionally priorities capital
expenditures and system maintenance based on solving system problems and
handling system growth. These expenditures are also related to maintaining an
acceptable level of reliability. In this period of deregulation such an objective could
be particularly important. Understanding the impacts of PQ variations on customers
requires monitoring of PQ along with customer’s follow-ups to assess the impacts.
• Identifying problem conditions: PQ monitoring can identify problem conditions
throughout the system before they cause widespread customer complaints, equipment
malfunctions and even equipment failures. (e.g. resonance conditions, switching
transients etc)

Some of the applications of intelligent PQ monitoring include

Industrial PQ monitoring:
• Energy & demand profiling with identification of opportunities for energy savings
• Harmonic evaluation to identify transformer loading concerns, resonance
concerns, mal-operation of equipment
 • Voltage sag evaluation to identify sensitive equipment and possible opportunities
for ride through solutions
• Power factor correction evaluation to identify proper operation of capacitor banks,
resonance conditions, etc.
• Motor starting evaluation to identify switching problems, inrush current concerns
etc.
Power system performance assessment
• Equipment performance trends, correlation with system performance trends and
identification of conditions that may need attention
• Voltage sag characterizing and assessment to identify the cause of voltage sags
and examining mitigation techniques.
• Capacitor switching monitoring to characterize switching transients.
•  Capacitor performance assessment
• Transformer loading for loss of life, harmonic-loading impacts etc.
• Feeder load monitoring and predictive assessment of future loads.

5. AUTOMATIC DISTURBANCE RECOGNITION SYSTEM (ADRS):

The work performed at the University of Tasmania in the past and
described above was done by analyzing the data manually and by visual inspection of the
waveform. This integration gives rise to an automatic disturbance recognition system
(ADRS). The ADRS will receive data online, process them and will be able to determine
whether there is a PQ disturbance in the particular group of data. In the event that there is
a disturbance, the system will also be able to classify the type of disturbance. The
components of the ADRS are shown in Figure 3.
Modeling for the ADRS
In order to train the Automatic Disturbance Recognition System, a large
database of disturbance waveforms is required. There are two possible sources for these
waveforms. The preferable source of disturbance data is from recording actual
disturbances in the power system. Suitable models must be constructed for each category
of disturbance the ADRS is to recognize.
6. ART TECHNIQUES: NEXT TRENDS
This section attempts to give a glance at signal processing techniques used in power
quality events classification.
A. Basic tools for signal components estimation.
Although root mean square (rms) is not an inherent signal processing technique, is the
most used tool. A great advantage of this algorithm is its simplicity, speed of calculation
and less requirement of memory.
B. What is next?
Hereto, a glance of the state of the art techniques oriented to automatic classification of
power systems events has been shown. Note that once data acquisition is made, signals
are band limited, which means that the information contained in the signal is trustable
until the Nyquist frequency (half the sampling frequency), e.g. if the sampling frequency
is 3600 Hz, the higher frequency component of the signal is 1800 Hz. Higher frequency
transients and traveling waves can hardly be studied from these signals.
7. SOME RESULTS
The system was tested with data containing known disturbances. The percentages of
successful recognition of the various types of disturbances are listed in Table 1.

It may be noticed in this table that for different types of disturbances, the success rate
varies from 93 to 100 percent except for the noise. This may further be improved if the
data is further processed. Work is presently going on towards this direction. The
relatively poor success rate in the recognition of noise is due the low sampling frequency
adopted in capturing the distorted waveforms. It is expected that with the increase of
sampling rate, the success will also improve.

8. CONCLUSIONS
The problems of power quality deviation have been discussed in this paper. Some results
of measurements in the Tasmanian power system have been mentioned. It brief
description of the automatic disturbance recognition system has been provided. Some
work has already been done on this and further work is going on. Results from
preliminary works with simulated data have been very encouraging. Sample results have
been provided in this paper. It is expected that with further research, the ADRS will
become a tool for near-perfect identification of all types of power quality disturbances.

9. REFERENCES

1. Trends in power quality monitoring, Mark McGranaghan, IEEE power engineering

review, October 2001.
2. An integrated approach to power quality improvement, R Venkatesh & S R Kannan, -
ET power tech 2001.
3. Power quality issues a distribution company perspective, IEE power engineering
journal, April 2001
4. http://www.washingtonpost.com
5. http://www.google.com

Document By
SANTOSH BHARADWAJ REDDY
Email: help@matlabcodes.com
Engineeringpapers.blogspot.com
More Papers and Presentations available on above site